Explanation When ingested orally, ascorbic acid is tightly controlled by four physiologic mechanisms: absorption, tissue transport, renal reabsorption/excretion, and utilization. Intravenous administration of ascorbic acid bypasses tight control until the kidney restores homeostasis, as determined from bioavailability experiments conducted by this laboratory. These data have surprising and novel implications for cancer treatment. More than 30 years ago, Ewan Cameron proposed that ascorbic acid might have a beneficial effect in treating patients with cancer. Joined by the two-time Nobel Laureate Linus Pauling, they published two case series indicating potential benefit of a large daily dose of ascorbic acid, 10 grams, in some patients with terminal cancer. Note for comparison purposes that the recommended dietary allowance for ascorbic acid at that time was 60 mg daily, or 0.6% of the treating dose. The Cameron-Pauling data were criticized because they were retrospective, without placebo control, in part subjective, and lacked independent pathologic confirmation. Investigators at the Mayo Clinic conducted two double-blind, placebo-controlled trials using 10 grams ascorbate daily to treat patients with advanced cancer, and found no effect. Based on these data, physicians were strongly advised to not use ascorbic acid in cancer treatment. In light of our extensive bioavailability data, we reviewed the experiments of Cameron and colleagues, and the Mayo Clinic investigators. We recognized that the Cameron-associated patients received IV and oral ascorbic acid, but the Mayo Clinic-associated patients received only oral ascorbic acid. Thus, comparisons between the treatment groups were invalid, and the issue of ascorbic acid in cancer treatment needed re-evaluation. Based on our bioavailability data, pharmacokinetics modeling, and measurements in clinical samples, intravenous ascorbic acid dosing can produce plasma concentrations as much as 100-fold higher than maximally tolerated oral doses. In vitro, pharmacologic ascorbic acid concentrations that are easily achieved in humans kill cancer but normal cells. Killing is mediated by extracellular ascorbic acid, its oxidation to ascorbate radical, and protein-dependent formation of hydrogen peroxide. To detect hydrogen peroxide without ascorbate interference, experiments required specialized chemical synthesis of peroxyxanthones, not commercially available. Based on the obtained data, we proposed and validated in vivo the hypothesis that pharmacologic ascorbic acid is a pro-drug for preferential formation of hydrogen peroxide in the extravascular space, but not blood. From these findings, we developed and continue to test a global hypothesis explaining why cancer cells but not normal cells are sensitive to ascorbate mediated death via extracellular hydrogen peroxide formation. Hydrogen peroxide in the presence of ascorbate and catalytic metals leads to formation of multiple reactive oxygen species (ROS). Reactive oxygen species thus formed may selectively kill cancer cells by one of many mechanisms: activation of poly ADP-ribose polymerase (PARP) and/or depletion of intracellular ATP; generation of DNA damage that cannot be repaired quickly enough to prevent cell death; depletion of intracellular reducing equivalents via their consumption by glutathione-dependent peroxidases; direct mitochondrial toxicity; direct membrane toxicity. Additional mechanisms are predicted and likely, all based on generation of extracellular and intracellular reactive oxygen species. Experiments exploring these concepts continue. Other current efforts are focused on: continued investigation of pharmacologic ascorbic acid as an agent to inhibit tumor growth in animal models with select treatment modalities; in vivo measurement of hydrogen peroxide in humans; initiation and continuation of phase II clinical trials; use of pharmacologic ascorbic acid in other diseases such as sepsis; and continued characterization of clinical usage patterns and safety. Phase I and Phase II clinical trials are promising, as are safety data to date.